ARCHIVED - Expenditure Review of Federal Public Sector - Volume Two - Compensation Snapshop and Historical Perspective, 1990 to 2003

• Previous Page
• Table of Contents
• Next Page

This page has been archived.

5. Financing increases in the Total Salary Mass

While we can determine, from amounts reported for Personnel expenditures in the Public Accounts annually, how much payroll and total compensation have risen in the core public service and separate employers for the period under review, there is no reliable method or data available to determine from what sources rising remuneration costs are financed. In this short chapter, we identify some sources of financing for the increased population and higher average salary witnessed in recent years, and attempt to quantify the amount dedicated to this purpose from each source.

Sources of compensation financing

Salaries in the core public service and separate employer domains totalled $8.2 billion in 1997–98, and rose to $12.4 billion in 2002–03, an increase of $4.2 billion. In general, the salary mass increase came from three sources:

• amounts approved by the Treasury Board for salary costs in support of new policies or workload increases;
• internal transfers by departments from non-salary budgets to salaries; and
• Treasury Board transfers to departments to cover salary increases from collective bargaining or other approved salary rate revisions.

Figure 2042 illustrates our conclusions about the relative size of the sources of increased salary funding.

We must emphasize that each dollar figure we present includes a significant margin of error. Thousands of transactions have been assessed electronically in an attempt to disaggregate the components. Errors can easily intrude that could alter the value of a component by as much as $200 to $300 million—say a margin of error of as much as 10% to 15%. We persist in presenting the results of our analysis despite the underlying technical fragility because the concepts are sound and the broad relativity among the components is useful to report.

Figure 2042
Estimate of sources of financing for the increase in the total salary mass for the combined core public service and separate employer domains, 2002–03 versus 1997–98

Display full size graphic

Important note:
All of the numbers in this Figure are estimates based on complex and (in-part) inter-related calculations relating to thousands of transactions. The actual total salary mass change could easily be within ± $200 million (that is ±5%) of the estimated $4.2 billion increase. At least as large degrees of uncertainty apply to all the components of the model. The $1.5 to $1.8 billion total for TB transfers for salary increases, for example contains some ambiguous data. The important point is that the model provides a useful overview of how salary increases were financed.

Program and policy changes financed by Treasury Board

We begin by highlighting increases in salary funding that resulted from Cabinet decisions to invest more in people in order to achieve public policy goals. As we explained earlier in this section, Treasury Board^[53] approved various proposals to fund new or expanded programs or to compensate for workload increases or other program integrity pressures on existing programs. After describing in some detail the approvals relating to five departments or agencies that were among the fastest growing between 1997–98 and 2002–03, we estimated that the total net additional salary funding approved by the Treasury Board for new policies or workload pressures in this period was around $1.3 to $1.6 billion.^[54]

This amount consisted mainly of additional amounts included in departmental Estimates through the Treasury Board Secretariat's Annual Reference Level Update (ARLU). This funding is part of the Main Estimates normally tabled in February or March each year. Also included is a small net amount relating to policy or workload approvals that were approved as part of Supplementary Estimates, or transferred to departments from Treasury Board votes that were part of the Main Estimates.

We must emphasize that constructing this story is based on several years of complex records, with dozens of decisions affecting most departments, and with increases and decreases interwoven in balancing plans against realities. Essentially this means that we must accept that our estimate is rough. It appears reasonable, however, to conclude that in the order of one third of the increase in salary mass between 1997–98 and 2002–03 resulted from decisions by Ministers to invest in additional people to achieve policy or program goals.

Budgetary transfers initiated by departments themselves

The second factor we note is the increase in net transfers into salaries from other approved budgets, for non-salary operating dollars, capital, and grants and contributions.^[55] The total net value of such transfers in the combined core public service and separate employer domains, comparing 2002–03 with 1997–98, was approximately $1.2 billion. This aggregates our estimates for in-year transfers, of around $865 million, and for permanent transfers listed in the ARLU, of about $315 million.

As we noted earlier in this section, these transfers could be financing:

• growth in the number of public servants, or
• increased salary costs arising from:
  • changes in the mix of occupations and classification levels in departments, or from
  • incomplete funding of salary increases agreed through collective bargaining.

We conclude that these transfers played a role in paying for both types of salary mass increases.

To gain perspective on how much of these transfers financed growing numbers of employees versus growing average salary costs, we evaluated the extent to which the policy/workload approvals identified already would have financed the growth in the number of federal public servants. The population of the public service expanded by about 37,000 between 1997–98 and 2002–03. Dividing our estimate of $1.3 to $1.6 billion for policy/workload increases approved and paid for by Treasury Board, by the population increase, yields an average salary of about $35,100 to about $43,200. This is clearly much too low, in that the average salary in current dollars grew from about $41,500 in 1997–98 to just over $52,800 at the end of 2002–03. This means that Treasury Board financing for policy/workload increases was insufficient to pay for the entire increase in the public service population.

We therefore multiplied the increase in the public service population (37,000 people) by the average of the annual average salaries over that period, which is about $47,500.^[56] This allowed for the fact that average salaries rose year over year during the period. This multiplication yielded a total cost of $1.8 billion. Thus we determined that, while departments received $1.3 to $1.6 billion in approved budgets for policy and workload increases financed by Treasury Board through ARLUs, they still had to find an additional $200 to $500 million to pay for compensation for the additional employees they hired.

This means that between one sixth and one third^[57] of the 37,000 increase in employees had to be paid for through in-year transfers effected by departments deciding, for example, that they could reduce travel or accommodation or program expenditures and hire more staff instead. These transfers, in turn, represent about 2% to 4% of the total 2002–03 salary mass.

Financing the average-salary increase

The final main component of the increase in salary mass is the increase in average salaries. As we have seen in the previous chapter, the average salary for the combined core public service and separate employer domains increased by $11,300—about 27.3% in current dollars—from around $41,500 in 1997–98 to about $52,800 at the end of 2002–03. (This corresponds to about 14.1% in constant 2002–03 dollars.) The factors contributing to this increase included:

• collective bargaining outcomes,
• both economic increases and changes to the structure of pay bands;
• a major shift in the composition of the public service toward more highly paid, more knowledge-intensive occupations;
• the net effect of annual increments within salary levels and departures from the public service; and
• settlements to implement equal pay for work of equal value.

To construct a gross estimate of the amount needed to finance this growth in average salary we made two calculations.

• First we multiplied the increase in average salaries, which is $11,300,^[58] by the 1997–98 population of about 198,000, giving a total of $2.24 billion.
• For the 37,000 employees added since 1997, we multiplied that population increase by $5,300, the difference between the 2002–03 average salary of $52,800 and the multi-year average of $47,500 referred to above. This calculation yields about $200 million.

Adding the two products gives a total cost of about $2.4 billion as the estimated amount needed to pay higher average salaries in 2002–03, compared with 1997–98.

If we are correct in our earlier estimate that $200 to $500 million of the net $1.2 billion of department-initiated transfers into salaries was needed to finance the growth in the public service population, this implies that the remainder (about $600 to $900 million) was applied to meet some of the cost of the increase in average salary. This suggests that an additional amount of about $1.5 to $1.8 billion was needed from the Treasury Board to pay for the approximately $2.4 billion increase in average salary. Because of the way records are maintained, we were not able to determine what amount was actually provided by the Treasury Board on a total net basis. One experienced analyst's educated estimate, however, was about $1.9 billion, a figure that lends support to our assumptions.

We must emphasize that these calculations are very rough. In particular, we acknowledge that separating the cost of hiring additional staff from the rising average cost of employees is somewhat artificial, in that a substantial portion of the change in the makeup of the workforce would have been effected through hiring additional staff. So hiring a new employee with more education or experience, at a higher-than-average classification level, would add both to employee population levels and to a higher average salary level. In consequence, our proposed split in the attribution of net departmental transfers between financing population growth and growth in average salaries is somewhat arbitrary. As we have seen, each element derives from assumptions. Nevertheless, the fact that the components more or less balance suggests that the model proposed is a reasonable approximation of reality. Such an understanding will assist in responsibly managing future increases in the public service salary mass.

How the federal government financed increasing compensation costs—1997–98 to 2002–03: a summary

We conclude that the $4.2 billion increase in the salary mass between 1997–98 and 2002–03, comprised three components, as follows.

Higher average salary—60 %

The cost of paying for higher average salaries amounted to about 60% of the salary mass increase.

• About three-fifths to three quarters of this was funded by the Treasury Board to cover the cost of collective agreements and other authorized pay increases.
• The remainder was funded from department-initiated transfers into salaries, mainly to cover part of the cost of changes in the mix of occupations and classification levels, as well as shortfalls in the amounts provided to cover collective agreements.

Approved staff increases for policy/workload purposes—about 33 %

About one third of the increase in total compensation costs resulted from Treasury Board approvals to hire more staff to deliver new policies or programs, or to deal with workload increases or other pressures on existing programs.

Department-initiated budget transfers for higher staffing levels—about 10%

The remaining cost of hiring new staff (more or less 10% of the total) was financed by department-initiated transfers into salaries from other approved budgets.

Having examined the macro forces and internal budgetary actions behind changes in the total salary mass, we now turn to an examination of specific aspects of compensation. We begin by looking at non-salary components of total compensation such as performance pay and other allowances. We then examine the federal government's pension plan and benefits programs for employees in the core public service and separate employer domains.

 

• Previous Page
• Table of Contents
• Next Page